In recent years, optical disk devices have attracted attention as means for recording/reproducing a large capacity of data, and are under active technical developments for achieving higher recording density.
Presently prevailing rewritable optical disks include spiral-shaped groove tracks composed of concave and convex portions (each having a width of about 50%) formed on a surface of a disk substrate at a pitch of 1 to 1.6 .mu.m. On the surface of the substrate, a thin film including a recording material (e.g., Ge, Sb, and Te in the case of a phase-change type optical disk) as a component is formed by a method such as sputtering. The disk substrate is fabricated in the following manner. First, a stamper is produced from a prototype where concave grooves and pits for sector addresses and the like are formed by cutting by light beam irradiation. Using such a stamper, the disk substrates made of polycarbonate and the like are mass-produced. The rewritable optical disks require sector-unit management for data recording and reproduction. Accordingly, at the fabrication of the disks, concave and convex portions (pits) are often formed on a recording surface, simultaneously with the formation of guide grooves for tracking control, so as to record address information of each sector.
Each track of the optical disk with the above structure is irradiated with a light beam having a predetermined recording power, so as to form recording marks on the recording thin film. The portions irradiated with the light beam (the recording marks) have different optical characteristics (reflection characteristics) from the other portions of the recording thin film. Thus, the recorded information can be reproduced by irradiating the track with a predetermined reproduction power and detecting light reflected from the recording film.
In the following description, the pits of physical concave and convex portions and the recording marks obtained by a change in the optical characteristics of the recording thin film are generically referred to as "marks", unless otherwise specified. The pits are read-only marks once formed, while the recording marks are rewritable. At the reproduction of recorded information, the two types of marks are read as changes in the amplitude of reproduction signals. The concave and convex portions as used herein refers to the shapes as are viewed from a reproduction beginning of the optical disk device. In other words, the "pits" refer to the convex portions as are viewed from the reproduction head, and the "grooves" also refer to the convex portions.
Techniques for achieving an optical disk with a high recording density include increasing the recording density in the track direction and increasing the recording density in the linear velocity direction.
Increasing the recording density in the track direction includes reducing the distance between tracks (the track pitch). One technique for reducing the track pitch is land/groove recording where signals are recorded both on convex tracks (groove portions) and concave tracks (land portions). The land/groove recording realizes double recording density, compared with the case of recording signals on either the groove porions or the land portions, if the other conditions are the same.
One technique for increasing the recording density in the linear velocity direction is referred to as mark length recording where both ends of a mark are made to correspond to "1" of modulation data. FIG. 1 illustrates an example of the mark length recording in comparison with inter-mark recording. Referring to FIG. 1, a sequence Y represents digital data modulated using a run length limit code. The run length limit code as used herein refers to a code sequence where the number of continuous "0"s interposed between every adjacent "1"s (hereinbelow, called the zero run) is limited to a predetermined number. The interval (length) from one "1" to the next "1" in the sequence Y is called an inversion interval. The limits, i.e., the minimum and maximum values of the inversion interval of the sequence Y are determined by the limitation of the zero run. Such values are called the minimum inversion interval and the maximum inversion interval.
When the sequence Y is recorded using the intermark recording (PPM; pit position modulation), the "1" of the sequence Y corresponds to a recording mark 101, and the zero run corresponds to a space 102. When the sequence Y is recorded using the mark length recording (PWM; pulse width modulation), the recording state, i.e., whether the recording mark 101 or the space 102, is switched by the appearance of "1" in the sequence Y. When the mark length recording is employed, the inversion interval corresponds to the length of the recording mark 101 or the space 102.
When a run length limit code of which the minimum inversion interval is 2 or more is used, the mark length recording may have an increased number of bits per unit length, compared with the inter-mark recording. For example, consider the case where the minimum value of the physical size of a mark which can be formed on a disk (called a mark unit) is the same in both the mark length recording and the inter-mark recording. As is observed from FIG. 1, while the inter-mark recording utilizes three mark units to record data of the minimum code length (three bits, "100", in the sequence Y), the mark length recording utilizes only one mark unit. For example, while the recording density in the inter-mark recording is approximately 0.8 to 1.0 .mu.m/bit, the recording density in the mark length recording is approximately 0.4 .mu.m/bit.
In general, the tracks on the optical disk are divided into recording sectors which represent minimum access units. Address information is prerecorded on each recording sector as described above. By reading the address information, the access to the recording sectors for data recording/reproduction is possible.
FIG. 2A illustrates a signal format of each recording sector of a rewritable optical disk which is in accordance with ISO (see ISO/IEC 10090). A recording sector 103 begins with a header 104 where addressing information for reading address information is prerecorded by forming concave and convex portions on the recording surface. A recording field 105 stores user data where digital data is modulated using a (2,7) modulation code for the inter-mark recording. FIG. 3 shows a conversion table of (2,7) modulation codes. As is observed from FIG. 3, by the (2,7) modulation, i-bit digital data (i=2, 3, 4) is converted into a 2.times.i-bit code sequence. The (2,7) modulation codes are run length limit codes where the zero run is limited between 2 and 7.
FIG. 2B shows the construction of the header 104. A sector mark SM is provided so that the optical disk device can identify the beginning of the recording sector without clock reproduction by a phase locked loop (PLL). As shown in FIG. 2C, the sector mark SM includes a pattern using comparatively long marks. Since the sector mark SM has this predetermined pattern, and the amplitude of the reproduction signals thereof is large, the sector mark SM is distinguishable from other data recorded using the inter-mark recording. The position of the header 104 is detected by detecting the sector mark SM, thereby to reproduce the address information.
VFO regions VF01 and VF02 shown in FIG. 2B are provided so that the optical disk device can obtain bit synchronization of reproduction signals using a clock reproduction by the PLL. A 2-zero run sequential pattern is recorded using the inter-mark recording.
Address marks AM are provided so that the optical disk device can identify the byte synchronization of subsequent address fields ID1, ID2, and ID3. Each of the address marks AM includes a pattern as shown in FIG. 2D recorded using the inter-mark recording technique. The pattern of the address mark AM includes a pattern of T.sub.max +1=9 bits where T.sub.max is a maximum inversion interval of the (2,7) modulation code (T.sub.max =8). This pattern does not appear in data recorded by the (2,7) modulation code.
Each of the address fields ID1, ID2, and ID3 includes: address information composed of track numbers, sector numbers, and the like; and cyclic redundancy check (CRC) codes for error detection during data reproduction, which are subjected to the (2,7) modulation and recorded using the inter-mark recording.
A postamble PA is provided to indicate the end of the (2,7) modulated data in the address field ID3.
FIG. 4 shows an example of signal amplitudes obtained when information recorded on the header 104 is reproduced by the optical disk device. As is observed from FIG. 4, the amplitudes of the reproduced signals are proportional to the lengths of the corresponding marks. The amplitude of the reproduced signal of the sector mark SM which has a long length is larger than that of the reproduced signal of other data. This allows for the identification of the sector mark SM by detecting the envelope of the reproduced signal waveform, and thus the detection of the beginning of each recording sector.
In the above example, all of the (2,7) modulated data is recorded using the inter-mark recording. However, in an optical disk having the header 104, when data is recorded using the mark length recording for improving the recording density, the marks recorded in the address fields ID1 to ID3 of the header 104 and the marks recorded in the recording field 105 have a certain length determined by the zero run limitation of the modulation code. Accordingly, the amplitude of the reproduced signal of data recorded using the mark length recording becomes large, compared with that recorded using the inter-mark recording where each mark corresponds to the 1-bit long "1". In the mark length recording, therefore, the difference in the signal amplitude (or the difference in the pattern) between the sector mark SM and the other portions becomes small compared with the case of the inter-mark recording. This makes it difficult to detect the beginning of the recording sector 103 by the envelope.
Moreover, when the above-described address mark AM is used, an erroneous detection of the address mark AM due to an erroneous bit shift of "1" may occur. For example, a code sequence obtained by the (2,7) modulation of digital data { . . . 10110011 . . . } is converted into { . . . 0100100000001000 . . . } from a conversion table such as that shown in FIG. 3. At this time, the pattern of the address mark AM is {0100100000000100} as shown in FIG. 2D. If "1" of the above (2,7) modulated pattern shifts by one bit, the resultant pattern is identical to the address pattern AM, which will cause erroneous detection.
In view of the foregoing, the objects of the present invention are to provide an optical disk, an optical disk device, and an optical disk reproduction method, where address information can be read reliably even when high recording density is achieved by employing mark length recording and the like.